Django’s database layer provides various ways to help developers get the most out of their databases. This document gathers together links to the relevant documentation, and adds various tips, organized under a number of headings that outline the steps to take when attempting to optimize your database usage.
As general programming practice, this goes without saying. Find out what queries you are doing and what they are costing you. You may also want to use an external project like django-debug-toolbar, or a tool that monitors your database directly.
Remember that you may be optimizing for speed or memory or both, depending on your requirements. Sometimes optimizing for one will be detrimental to the other, but sometimes they will help each other. Also, work that is done by the database process might not have the same cost (to you) as the same amount of work done in your Python process. It is up to you to decide what your priorities are, where the balance must lie, and profile all of these as required since this will depend on your application and server.
With everything that follows, remember to profile after every change to ensure that the change is a benefit, and a big enough benefit given the decrease in readability of your code. All of the suggestions below come with the caveat that in your circumstances the general principle might not apply, or might even be reversed.
…including:
Field.db_index
or
Meta.index_together
to add
these from Django. Consider adding indexes to fields that you frequently
query using filter()
,
exclude()
,
order_by()
, etc. as indexes may help
to speed up lookups. Note that determining the best indexes is a complex
database-dependent topic that will depend on your particular application.
The overhead of maintaining an index may outweigh any gains in query speed.We will assume you have done the obvious things above. The rest of this document focuses on how to use Django in such a way that you are not doing unnecessary work. This document also does not address other optimization techniques that apply to all expensive operations, such as general purpose caching.
QuerySet
s¶Understanding QuerySets is vital to getting good performance with simple code. In particular:
QuerySet
evaluation¶To avoid performance problems, it is important to understand:
As well as caching of the whole QuerySet
, there is caching of the result of
attributes on ORM objects. In general, attributes that are not callable will be
cached. For example, assuming the example Weblog models:
>>> entry = Entry.objects.get(id=1)
>>> entry.blog # Blog object is retrieved at this point
>>> entry.blog # cached version, no DB access
But in general, callable attributes cause DB lookups every time:
>>> entry = Entry.objects.get(id=1)
>>> entry.authors.all() # query performed
>>> entry.authors.all() # query performed again
Be careful when reading template code - the template system does not allow use of parentheses, but will call callables automatically, hiding the above distinction.
Be careful with your own custom properties - it is up to you to implement
caching when required, for example using the
cached_property
decorator.
with
template tag¶To make use of the caching behavior of QuerySet
, you may need to use the
with
template tag.
iterator()
¶When you have a lot of objects, the caching behavior of the QuerySet
can
cause a large amount of memory to be used. In this case,
iterator()
may help.
For instance:
F expressions
to filter
based on other fields within the same model.If these aren’t enough to generate the SQL you need:
RawSQL
¶A less portable but more powerful method is the
RawSQL
expression, which allows some SQL
to be explicitly added to the query. If that still isn’t powerful enough:
Write your own custom SQL to retrieve data or populate models. Use django.db.connection.queries
to find out what Django
is writing for you and start from there.
There are two reasons to use a column with
unique
or
db_index
when using
get()
to retrieve individual objects.
First, the query will be quicker because of the underlying database index.
Also, the query could run much slower if multiple objects match the lookup;
having a unique constraint on the column guarantees this will never happen.
So using the example Weblog models:
>>> entry = Entry.objects.get(id=10)
will be quicker than:
>>> entry = Entry.objects.get(headline="News Item Title")
because id
is indexed by the database and is guaranteed to be unique.
Doing the following is potentially quite slow:
>>> entry = Entry.objects.get(headline__startswith="News")
First of all, headline
is not indexed, which will make the underlying
database fetch slower.
Second, the lookup doesn’t guarantee that only one object will be returned. If the query matches more than one object, it will retrieve and transfer all of them from the database. This penalty could be substantial if hundreds or thousands of records are returned. The penalty will be compounded if the database lives on a separate server, where network overhead and latency also play a factor.
Hitting the database multiple times for different parts of a single ‘set’ of data that you will need all parts of is, in general, less efficient than retrieving it all in one query. This is particularly important if you have a query that is executed in a loop, and could therefore end up doing many database queries, when only one was needed. So:
QuerySet.values()
and values_list()
¶When you just want a dict
or list
of values, and don’t need ORM model
objects, make appropriate usage of
values()
.
These can be useful for replacing model objects in template code - as long as
the dicts you supply have the same attributes as those used in the template,
you are fine.
QuerySet.defer()
and only()
¶Use defer()
and
only()
if there are database columns
you know that you won’t need (or won’t need in most cases) to avoid loading
them. Note that if you do use them, the ORM will have to go and get them in
a separate query, making this a pessimization if you use it inappropriately.
Also, be aware that there is some (small extra) overhead incurred inside
Django when constructing a model with deferred fields. Don’t be too aggressive
in deferring fields without profiling as the database has to read most of the
non-text, non-VARCHAR data from the disk for a single row in the results, even
if it ends up only using a few columns. The defer()
and only()
methods
are most useful when you can avoid loading a lot of text data or for fields
that might take a lot of processing to convert back to Python. As always,
profile first, then optimize.
QuerySet.count()
¶…if you only want the count, rather than doing len(queryset)
.
QuerySet.exists()
¶…if you only want to find out if at least one result exists, rather than if
queryset
.
But:
count()
and exists()
¶If you are going to need other data from the QuerySet, just evaluate it.
For example, assuming an Email model that has a body
attribute and a
many-to-many relation to User, the following template code is optimal:
{% if display_inbox %}
{% with emails=user.emails.all %}
{% if emails %}
<p>You have {{ emails|length }} email(s)</p>
{% for email in emails %}
<p>{{ email.body }}</p>
{% endfor %}
{% else %}
<p>No messages today.</p>
{% endif %}
{% endwith %}
{% endif %}
It is optimal because:
with
means that we store user.emails.all
in a variable
for later use, allowing its cache to be re-used.{% if emails %}
causes QuerySet.__bool__()
to be called,
which causes the user.emails.all()
query to be run on the database, and
at the least the first line to be turned into an ORM object. If there aren’t
any results, it will return False, otherwise True.{{ emails|length }}
calls QuerySet.__len__()
, filling
out the rest of the cache without doing another query.for
loop iterates over the already filled cache.In total, this code does either one or zero database queries. The only
deliberate optimization performed is the use of the with
tag. Using
QuerySet.exists()
or QuerySet.count()
at any point would cause
additional queries.
QuerySet.update()
and delete()
¶Rather than retrieve a load of objects, set some values, and save them individual, use a bulk SQL UPDATE statement, via QuerySet.update(). Similarly, do bulk deletes where possible.
Note, however, that these bulk update methods cannot call the save()
or
delete()
methods of individual instances, which means that any custom
behavior you have added for these methods will not be executed, including
anything driven from the normal database object signals.
If you only need a foreign key value, use the foreign key value that is already on the object you’ve got, rather than getting the whole related object and taking its primary key. i.e. do:
entry.blog_id
instead of:
entry.blog.id
Ordering is not free; each field to order by is an operation the database must
perform. If a model has a default ordering (Meta.ordering
) and you don’t need it, remove
it on a QuerySet
by calling
order_by()
with no parameters.
Adding an index to your database may help to improve ordering performance.
When creating objects, where possible, use the
bulk_create()
method to reduce the
number of SQL queries. For example:
Entry.objects.bulk_create([
Entry(headline='This is a test'),
Entry(headline='This is only a test'),
])
…is preferable to:
Entry.objects.create(headline='This is a test')
Entry.objects.create(headline='This is only a test')
Note that there are a number of caveats to this method
, so make sure it’s appropriate
for your use case.
This also applies to ManyToManyFields
, so doing:
my_band.members.add(me, my_friend)
…is preferable to:
my_band.members.add(me)
my_band.members.add(my_friend)
…where Bands
and Artists
have a many-to-many relationship.
Jun 22, 2017